US 11,922,217 B2
Systems and methods of optimizing resource allocation using machine learning and predictive control
Shihui Chen, Boston, MA (US); Keon Shik Kim, Cambridge, MA (US); and Douglas Hamilton, Boston, MA (US)
Assigned to Nasdaq, Inc., New York, NY (US)
Filed by Nasdaq, Inc., New York, NY (US)
Filed on Nov. 13, 2020, as Appl. No. 17/097,178.
Prior Publication US 2022/0156117 A1, May 19, 2022
Int. Cl. G06F 9/50 (2006.01); G06N 20/00 (2019.01)
CPC G06F 9/5027 (2013.01) [G06F 9/50 (2013.01); G06N 20/00 (2019.01)] 21 Claims
OG exemplary drawing
 
11. A method, comprising a computer system that includes at least one memory and at least one hardware processor:
receiving by a transceiver over a data communications network different types of input data from multiple source nodes communicating with the data communications network;
executing, by a processing system that includes at least one processor, instructions stored in memory as follows:
(a) defining for each of multiple data categories, a set of groups of data objects for the data category based on the different types of input data;
(b) predicting, using one or more predictive machine learning models, a selection score for each group of data objects in the set of groups of data objects for the data category for a predetermined time period;
(c) determining, using one or more control machine learning models, a number for each group of data objects indicating how many data objects are permitted for each group of data objects based on the selection score for each group of data objects;
(d) prioritizing, using one or more decision-making machine learning models, the permitted data objects based on one or more predetermined priority criteria;
(e) monitoring activities of the computer system for data objects actually selected during the predetermined time period;
(f) calculating performance metrics for each group of data objects predicted to be selected during the predetermined time period based on the data objects actually selected during the predetermined time period; and
(g) adjusting the one or more predictive machine learning models, the one or more control machine learning models, and the one or more decision-making machine learning models based on the performance metrics to improve their respective performances.